I'm tempted to say this is true (most newer developers are unlikely to have a deep understanding of older, faster languages like C++, especially in the age of Javascript and glorified IRC clients that consume more resources than the OS does to run), but I think a more accurate description of the problem is that Windows became fragmented in 2012 for stupid reasons (when the decision was made to make desktop Windows an ad for its unfinished tablet version- Metro was an unmitigated usability disaster for everything not a tablet) and everyone getting hired on is just trying to put the pieces back together.

We don't know if Windows is in maintenance mode or not either, or what the next version would look like if one eventually comes out (maybe it'll all be in TypeScript?). After all, UWP was mostly retired and the visions of "your next desktop computer is a tablet" have significantly died down- the technology and the will of the market just weren't there.

However, Microsoft has failed once before to develop a proper, stable OS- though Vista did bring a great number of graphical and usability updates to XP, chief among them the "Show Desktop" button, the ability to snap windows onto sides of the screen, and putting a Firefox-style search bar in Explorer, it suffered from poor functional quality (file copy windows should never ever be broken, for instance), so I'm not convinced it's solely the fault of newer developers.

I wouldn't blame the new people. You can't possibly keep up with a huge code base as Windows when you can't implement anything smart as it can always break something somewhere. You need huge tests just to change a small thing.

It would give them a better return to just make a new OS from scratch while keep supporting Windows for the next 10 years (maybe 20 for extended support) and let it slowly fade.

Or better yet, take the Edge route and use Linux underneath and change the appearance to MS' taste, have binary compatibility layer and call it Windows.