The issue I have with the "D&D as DIY hobby" viewpoint approach is this:
Why would I ever DIY for my home game using D&D as the foundation, when FATE and Cortex are right there and come with much less baggage to unpack and discard?
If I was in the business of producing RPGs commercially, then I could see the argument for making a 5e hack along the lines of AiME, Esper Genesis, Five Torches Deep, Pugmire, etc. Alternatively, I could try and make something new out of the rules chassis of earlier editions; Pathfinder did it with 3.5, Lancer did it with 4e, and the OSR market is (over)saturated with B/X and AD&D clones and hacks. But if you aren't interested in going so far as to creating a new game entirely (and possibly monetizing it), then I don't see trying to make 5e work for purposes beyond its RAW state worth the effort.
In the context of the 90s and the early 00s, when the Internet wasn't as developed as it is today and both games themselves and in-depth discussions on games theory and design wasn't as accessible, I can understand why DIY became the dominant ethos. But the field has shifted, the breadth of tabletop games has massively expanded, and I am very tempted to say that the community overall has a better understanding of the hows and whys of the ways tabletop RPGs function.
In that light, I feel that there's two ways you can go with DIYing 5e and have it be worth the time investment: you either make some very minor aesthetic changes, or commit to really taking it apart and putting it back together in a standalone game. If you don't commit to either extreme, you'll more than likely find yourself in some weird limbo of messy and clunky design that wasn't really worth the effort and probably would have been better off just using another game for your purposes, whether generic or specific.