There's also something here which Microsoft used to know in the 9x days and has since forgot (because it's not immediately profitable to remember in the current paradigm). The needs of a single-user system are different from the needs of a network terminal. Since Windows 2000, every Windows PC is a network terminal. Since OSX, every Apple PC is a network terminal. Unix-derived OSes have always been targeted at network terminals. In a network terminal, the machine might be used by more than one user; therefore, you can't allow too much customization. The data on that terminal might be needed by another user on the network; therefore you must protect a significant amount of that data and must prioritize system operation over user desire at every turn.
Oh, and one more thing. Stop worrying about profit and trying to hamstring the user. Either create a good, working, product that you stand behind with every fiber of your being... or don't create a product in the first place.
BS. Applications in scientific research easily blow that ridiculous limit out of the water. Your point about single vs multi user is good, but you do yourself no favors by starting off with a bad premise. And lest you say you're talking about personal OS only, that's not what you wrote at all.
Even in sciences where it shouldn't be the norm, we tend to have "dumb" people reliant on smart computers for interpretation of results, instead of smart people using dumb computers as tools to aid computation.
Frankly, even if you still believe there's reason that certain research scientists would need that kind of processing power and storage at times, that doesn't excuse the fact that the OS should not be taking up that time. An OS should exist to load user programs, allow data access by users and programs, provide a platform for peripheral connections as needed, and otherwise get out of the way with a bare minimum load.
Davis' work shows the way to a completely different path than the prevailing hegemonic trend to try to build this or that system that everyone will use, that will take over the world. Maybe everyone deserves their own completely idiosyncratic and personalized system that can keep on working for them despite whatever the industry behemoths do. Perhaps an interesting research project might be to try to find ways to enable ordinary developers or even ordinary people to make such personalized stand-alone systems with no dependencies, without requiring years of fanatical single-minded effort.
In his own words, he wanted a modern day Commodore 64. And God bless him for it!
Having said that, I think the two main areas that I've seen praised are the "coding the image" approach he took which has its predecessors in Lisp machines and Smalltalk. Also the pervasive hypertext and embedded media.
Of course, those things aren't especially original and it's not obvious to me that he took them in directions that are especially novel.
On a purely speculative level, what might catch on (though I give it a very small probability) is that TempleOS is a deliberately un-networked OS that gives complete control to the user. It's not inconceivable that, as paranoia becomes the sensible default, more folk will decide that that's a good place to be. I doubt it but you never know.
