It's really hard to maintain a product team where the mandate is just "don't break anything and keep the quality high". Especially something with as big of an installed base as Windows.
The team will look for excuses to build new and exciting stuff and new opportunities to increase revenue. Even if the product is pretty much "done".
No, it really shouldn't be... You can reduce headcount a lot, which they did, and concentrate on bugs (including security reports), while working with hardware vendors for if/when new features need to be integrated for better usability.
If/when you decide to do a redesign, it should be limited to a specific area, or done in such a way that all functionality gets moved to its' new UI/UX in a specified timeframe and released when done. Not, oh, here's a new right click menu that you now have an extra click 1/3 of the time for the old menu that has what you are actually looking for because the old extension interface was broken.
Want a real exercise in fun ... just for fun, because I know it's not as useful on a laptop, but was fun on desktops... get a screensaver working in windows that runs for an hour or so before going to sleep... just try it... that's a fun exercise in frustration... oh, it's still in there, but every third update will disable it all again. I get it... but you know what, I want my matrix screensaver to run when I'm only away for a few minutes or over lunch.
I disagree, I think companies mostly just don't want to spend development money on existing "finished" products. That's the smell I'm getting from microsoft.
There are plenty of easily identifiable issues with performance in windows 11. There should be people in the windows team dedicated to eliminating "jank". MS product owners, on the other hand, are much more interested in getting copilot integrations into every menu. That's an "easy" task which looks good on a scorecard when you complete it.