I think that's a misstatement; I'm fairly certain that AES works well as a composite function (with non-related keys), altho' I would personally use a second cipher for the second round. Even in the case of the 3DES MITM style attack, 3DES remained more resilient then just DES, just not 3x as effective.
I missed the part where you explained why it's a misstatement.

.
It's note barely useful, it's useful, just closer to 80% useful. Depends.
For znort987, I recommend a chain of multiple ciphers for composite encryption, and unrelated random keys.
If the user can secure the key of such system, he could as well secure the bitcoin key directly.
Depends on his setup. But I like the comic!