When learning, unlike the person teaching you, you don't know what's coming in the next lesson or the one after it, so it's sometimes a good idea to trust them on a curriculum. It's also the one of the biggest downsides of being self-thought (as I am in big part), you always have holes in your perspective that you're not even aware of and it takes years to stumble on something that someone else was told in the 1st year of being a dev.
Only if one knows what they don't know. And fundamental concepts have a habit of hiding in plain sight. It's easy to waste lots of resources because one doesn't know about a fundamental concept.
You can probably find a solution with what you already now (as a junior programmer) as most problems programmers have to solve are not that hard but you may completely miss a better solution because you had no idea it was possible. I may be fine but you may also lose a lot of time later because it wasn't.
I agree that it's possible to self teach almost everything in CS but the point of university is to speed up the discovery of CS from scratch and have solid foundations. You certainly don't know everything graduating university but should now where to look when you have a problem imo.
If you pass an earlier test while missing some fundamental concept, then you'll simply spend the time learning it when you do need it. If you never need it, maybe it wasn't that fundamental after all.