Educating humanity is difficult. Every solution has its flaws. Giving up on it is worse than every flawed solution, however.
Not everybody loves to be forced to be sitting in a school classroom for 7 hours a day, 5 days a week. And not everybody want to memorize state capitals.
There are 3 or so Billion people living in civilisation, oddly enough, have somewhat similar education systems.
With some degree of variation, there doesn't seem to be any magic, other than the things we know are going to work.
And yes, we know a good chunk of it is just daycare and that's ok too.
The identified problems remain, as do all of the original problems that the outgoing system has solved. The aeons of tradition that have built up to the current solution will have to simply be rediscovered the hard way, perhaps with different window-dressing, but ultimately not that far off from their previous incarnation, especially if the deconstruction happens by those who haven't truly learned the lessons about what went well and what went badly before.
Sometimes this comes about because the wishful solutions we want to apply don't work; sometimes because reality is messier than we want to admit, and the solutions are messier still. Child discipline, for example, may always end up being harsher than we as gentle, respectful people would like it to be, particularly for unruly children. Institutions may always be slower to react and improve than we would like.
Much like in software, a full rewrite is often only a pathway to a new set of problems, compared to a refactoring that could solve the known set of problems that remain without losing any existing value.