Perfection cannot be reached, but better is worth striving for, or we'd still be playing Chainmail.
Better isn't always actually "better", however. Its subjective and in game design certain routes taken can have a lot of unintended consequences.
For instance, making magic users easier to access and use is a good idea in general for a game compared to what we had before.
However, in the execution of that idea we ended up here with busted demigod mages that are, regardless of how severe we consider the problem, impacting the game negatively.
Its not uncommon for one of the basic bits of advice for homebrewing is to keep your changes small, because larger changes can introduce problems predictable and unpredictable.
If you jack up martials you
will have to completely rewrite all monsters; thats not going to be negotiable unless they decide to just foist all the balancing problems onto DMs to figure out. (Which means nothing actually changes)
And thats without even getting into fixing the non-combat game. Exploration is already devalued mechanically as it is and needs a serious overhaul. Jacked up martials are going to require that overhaul be even more drastic, and that in turn means
even more consequences can come up as a result.
Thats why a complete rewrite (new edition) is a more ideal way to address 5e's problems if you're not going to commit to invalidating what people already own. A new edition means 5e can just be its own thing, and you can start over and build up the system properly.
And of course it can't be ignored that for WOTC to take that action would be their third go-around with doing exactly that, and their track records leaves a lot to be desired for how that'd turn out.
But it doesn't change that one way or another, to fix 5e goes deeper than adding or changing a class or two. Way deeper, in fact.
Ive said it before but if all you did was add a new class or jacked up all the Martials, the disparity
would still fundamentally exist, because it isn't
only being caused by just class design.