Before dismissing it out of hand, like you are on the board of Apple - think for a second.
As both sets of silicon get more efficient and cooler - it would all work - share the same cooling architecture.
Why can't we have the best of both worlds?
Because you're comparing two different technology stacks and on an unbalanced factor. This has nothing to do with cooling at all and they're not even in the same group. You're comparing apples to oranges.
Apple Silicon is a complete system on a chip (SoC) with its own GPU, neural engine, ISP, disk controller, memory controller, cache, unified memory, co-accelerators, etc.
Intel CPU is just a x86 CPU and on some, basic iGPU. (I'm ignoring the certain instruction sets that it may have but Apple won't care much for).
Apple silicon is NOT CPU only. It's also not about ARMv8 either.
Apple is integrating both software and hardware on a much deeper levels that has never been possible before.
Intel does not produce custom x86 chips for Apple, they may "bin" certain pref chips but they do not customize their designs to anyone (unlike AMD).
Apple is customizing their own hardware for their own software, that is *always* going to be faster than a general CPU that's designed for mass market.
Imagine if they know that certain group of software is slow on certain tasks, they can customize the next SoC to accelerate that task that'd be much faster than trying to increase the clock speed.