Isn't CT much about higher-order parameterized type-signatures of a set of "classes" that can usefully work and be combined with each other? Then instead of gr

Isn't CT much about higher-order parameterized type-signatures of a set of "classes" that can usefully work and be combined with each other? Then instead of grokking the CT and its theorems as a whole it would seem to be useful to learn about such type-signatures among a set of collaborating (OOP-like) classes which are useful for many programming tasks? In a sense categories then could be seen as formalization of "Design Patterns"? Something like that?