For example, let's say you are trying to solve a problem in crystallography. You need to be a good programmer to come up with a nice refinement algorithm, but without a fundamental understanding of the problem you're trying to solve, you're not going to be able to solve it. The same goes for financial modeling, or some other kind of optimization problem.
In any field you have to keep teaching yourself new skills, since the technologies and approaches change. Even the foundational knowledge changes, but what education helps you get is a core understanding of depth that allows you to adjust. There will always be those who can do it without getting a good education, but those are rare, and in the worlds I've lived in, almost non-existent.
I think one of the large gaps between those groups on average is those without a CS degree think theory is useless. When you are forced to learn the theory you end up using it constantly. However, when you see a theory without the associated experience it seems useless. So it's hard to know what to study on your own. For most people it's the rigorous nature of a good CS program that forces them to put forth the associated effort.