To carry on with the automotive analogy, I've likely achieved "level 5" of a car: I've done several clutch replacements, rebuilt a suspension, and my living room is full of car parts turned art pieces. But there are still things I want to do that extend the functionality of the car, so it can do things the original designers never intended, but that suit my needs better.
I think this is the critical step in innovation: taking something as it is and being able to extend traditional "mastery" of the subject and create something new and valuable. At that point, the cycle starts over: most people won't use the thing, or some will use it for its first order purpose, but some dedicated group of folks will dive deep and build on what you did to create something new.
Educationally, I used to chat with my professors about how the theory they used to write tests, and the best ones always fixed 100% at something we had not yet learned, but could be derived from what we had been taught. You could still get an A (90%) by regurgitating information by rot memorization, but that extension was what showed you could move the field forward. Not only did it make the tests more challenging, it also made the tests more fun.