> 4e wasn't “discard[ing] a big chunk of its worst, thorniest rules”, it was a re-write to sell splat books and accessories
The two are not mutually exclusive: the former is game design, the latter is business aim motivating game design, product strategy, and lots of other things. It was both.
> 5e was the streamlining and modularization that was needed so you could play it like it was 2e, or 3e/3.5e, or even 4e if you wanted to.
Some modularization, sure, but it many ways it took small steps back from 4e's mechanical streamlining. Which is a good thing, 4e's extreme mechanical streamlining without regard to the role of the mechanics at the table in service of RP is a big part of why 4e ended up flavorless and dull. 5e kept most of the mechanical substance of that streamlining, but refocussed on serving RP and, in so doing, made some compromises to the mechanical streamlining. This made it more accessible for the same reason a lot of programming languages that have less refined, pure, coherent, generalizable abstractions than Haskell are more accessible.
(Also, I think you are doing the people working on business strategy at WotC a disservice if you think 5e is any less well designed to sell splat books and accessories than 4e. In 5e, the choices matter more to players -- which makes having more choices more valuable. And returning to OGL and adding DMs Guild means that there is more opportunity for third parties to supply the relatively low-margin long-tail supplements that each have a small market but collectively provide a strong ecosystem that keeps people buying the higher margin core books and major supplements and accessories that Wizards dutifully churns out.)