> Just using the same encryption algorithm twice with two different keys of
> size N does not increase the exhaustive search time from 2^n to 2^2n,
> thanks to meet-in-the-middle attacks, which reduce the time to 4^n.
I think there is a mistake somewhere in there, because 2^(2n) = (2^2)^n = 4^n.