Kurt Lewin once said just that, and I tend to agree. At least as long as we know the assumptions behind the theory. Because when we do, theory becomes a way of seeing. A filter that lets us see the world around us, and how it works, in a different light. In my teaching, a key goal is to help people see, with theory.
In doing so, I also help myself see. Feynman reportedly said that if you want to master something, you should teach it. I second that. Because when I try to carve something down to its essence to explain it to others, I clearly see the holes in my own logic, and the gaps in my knowledge I need to fill first.
Finally, I also see teaching itself as a creative activity. Where I can experiment with formats, test new ideas and, often, get new ones myself.