1) Apple shifting the Mac lineup into consumer and pro levels. Differentiating said products by use so that pro users have upgradability without constant machine upgrades. I just recently purchased a Mac Studio. I'm likely going to be very happy with the machine (M2 Max, 32 Gig RAM, 512 GiB drive) but I still after all these years yearn for a realization of the xMac (mythical beast talked on messageboards going back to the launch of the Cube and the Mac mini later). While I know many will mention that having the memory on the SoC is a performance advantage... I also know that very few people in computing are crying about how slow standard DDR5 memory is let alone how slow PCI-E Gen 3/4 NVME m.2 SSD's are. There is a time in all of this that Apple eating their own dog food becomes a bad thing. There is a time in all of this when some of Apple's antics, highlighted by people like Quinn from Snazzy Labs or Marques Brownlee (MKBHD) or even PC Master Race standout Linus Sebastian (Linus Tech Tips; who does have a Mac channel FWIW and does admit to liking Apple at times) are 100% right when they call Apple out on their ridiculousness. From the ridiculous price of wheels on a Mac Pro to the insane price of a stand for a Display XDR... Apple does their part to feed to that insanity and make it harder for a Mac fan (that isn't a fanboi) to call them on it.
Basically put... I was a Steve Jobs fan. I'm an Apple fanatic. I've owned Macs since my first 7100/66cd after moving over from my Amiga 500 I had prior. I don't ever regret the decision (though I regret getting rid of the A500, especially since they still sell for a premium and I miss It Came From the Desert). That said, as someone that took product design... that's watched probably every Apple-themed movie or read every Apple-themed book (and a few books of other tech people; shout out to Howard Scott Warshaw on my last read).
Jobs wasn't right on everything. He wasn't perfect. Even his movies have highlighted this. "I was made poorly." His hang up on not seeing the internals is part and parcel why he's not here with us, sadly. There's nothing wrong with having user serviceable and accessible designs. There's nothing wrong with Apple giving it's customers the ability to choose to sideload applications on a device (even if I likely won't choose that myself, and as long as IMHO it's not enabled by default nor able to be switched by a 3rd party application - ergo it MUST be done by the customer and not by someone like Epic Games, etc.). Finally, there's a point in all of this where the Bauhaus' influence on Apple products needs to come back to the reality that "form follows function" isn't always being adhered by Steve's and Jony's visions on designs. You CAN have an elegant design that much like the K2-series beige desktops/towers or the various iterations of Pro desktops dating back to the blue and white G3... where you pull a tab or push a button and the door opens or case just slides off. Heck, give me a mini and Studio design where there's a push button on the front and the whole thing opens clamshell? Give us the ability to flip it over and slide open a door to access NVME slots. All of this can be done and be something that's the envy of other PC users... not something that is laughable.
Finally... dual tier on the ARM processing front. Use the Cortex-based A#/M# series for iPhone, iPad, and consumer Macs. Give us a MP# series (Mac Pro) based on the Neoverse E-series. Basically put, a powerful yet efficient high end chip with more cores, SMT capabilities (think back to when PowerPC had AltiVec), and that are powerful enough to compete with Xeon and Epyc class workstations and yet efficient enough to go in a laptop that now can be a mobile workstation. I do believe that having dedicated memory for graphics is a performance enhancement that's not to be denied, but to me there's no reason we can't have DDR 5 for most memory applications and that in the here and now Apple should give us as much as possible with as easy an outlet to get to it as possible. With Right to Repair basically being a done deal, making this more feasible and common on the desktop and in laptops from Apple should go a long way towards appeasing us as customers. Just because we love the Mac, doesn't mean we have to love everything Apple does. Just because Jobs was a visionary and had a great sense of style... doesn't mean he was always right. And if Tim Cook can be the bridging point to what is right and the great things that Jobs brought to Apple, and yet still bring in some of the greatness of what Wozniak would still lobby for if he was still working there? I think Tim can go down in history as something greater than even Jobs, without desecrating anything Jobs brought to the table.
2) Software. I constantly feel like Apple makes great products and then kills them mid-stream for no reason. Pages, Numbers and Keynote had a lot of hype when they launched and then over time the hype meter just went away. Apple had a two-tier approach with Final Cut that they ultimately axed. I guess you can even call it 3-tier if you bring iMovie, Final Cut Express and Final Cut Pro into the mix. There's the reality that Apple similarly had iPhoto and Aperture... they let iPhoto become bloated and buggy and they just outright killed a beloved Apterture application that many Mac users didn't want to see go. There's the fact that Apple clinically still owns Claris who makes Filemaker, and yet you never even hear anything about the application. Basically put, I feel like Apple needs to spearhead a better team when it comes to software and allow that team or side of Apple to do greater things. That doesn't mean they have to step on other companies' software but it also doesn't mean that they can't come onto the scene and compete and drive the market.
3) Games. Now that we have the developer toolkit for getting DX titles running on the machine to assess how to get the games ported, we need not only for Apple to continue to double down on this tool... but to contribute to the open-source tool that they are literally using to make it work. To do their part to not only make it easier to get games ported to macOS but also using the fact that a version of the core software that can run on macOS could also be a stepping stone to iPadOS and iOS applications increasing the potential revenues on the efforts to port to Metal and the Apple API's. But, it also doesn't hurt if that same codebase also makes it easier to get games ported to Linux as well. Why? Because anything that can shrink Windows marketshare and make it a necessity to deploy on any and all capable systems is beneficial to each player. Apple couldn't make that toolkit without Crossover which has made their product for helping people deploy DX games on Mac and Linux. Making it a stepping stone to native games potentially makes it easier to get there.
That said, Apple needs the Mac to finally get marketshare. To get that, many people are going to want more compelling reasons to make that purchase. With NVidia and AMD seeming to focus their GPU developments in ways that are foregoing the gaming market in some capacities? You have an outlet for Apple to get more consumers to consider other choices the more their ARM computers progress. That said if you lock people out of upgrades to the point of constant new machine purchases... it might stifle that. So #1 could use improvement IMHO.
And ultimately, if you don't get more games, the "killer apps" that consumers want, you aren't likely to increase marketshare amongst consumers on the Mac and you aren't likely to incentivize developers to bother with a port regardless of how simplified the process might be or how much getting a iOS or iPadOS port can pave the way to a slam dunk easy macOS port. But if the code can work on iOS, iPadOS, the Mac and say... the popular Steam Deck. If Apple puts a greater emphasis on software (see #2) and #3 (adds a game development division while continuing to emphasize dev tools to help game developers outside of Apple) and continues to contribute to the very tool that made some of this possible and might help make both macOS and Linux eat into that marketshare and make it more important to write for them all and play to all audiences. That's a point where the Mac has it's duck in a row, where it's a product that those that own iPhones can appreciate for not only it's superior integration between the iPhone, iPad, Watch and Mac... but... where they'll be more keen to embrace that integration because it's not just a dull boy (all work and no play).