Depends if you only want to show that set theory is incomplete, you don't need Gödel numbering and you can more-or-less turn Smullyan's explanation into a complete proof in a straightforward manner.

You're right, I hadn't thought about that.