Nah, machine verification just tells you that a theorem is true. It does nothing to help you understand it. An Lean formalization of Wiles' proof of FLT would be as incomprehensible to me as Wiles' non-formalized journal article is.