Has there been a threat model performed by a security person (not a developer)? Functional testing only goes so far and stops at the assumption that the code a

Has there been a threat model performed by a security person (not a developer)? Functional testing only goes so far and stops at the assumption that the code and design are correct.