Others have specifically addressed where you went wrong.
The standard way of attacking the problem is as follows:
Let
$$\binom{n}{k} = \frac{n!}{[k!][(n-k)!]} ~\text{where}~
n \in \mathbb{Z^+}, ~k \in \{0,1,2,\cdots, n\} ~\text{and}~ 0! = 1.$$
Here, $\binom{n}{k}$ represents the number of different ways of selecting $k$ items from a group of $n$ items.
Let $A, B, C$ denote the # of different ways that Amy and Bunny can select shirts, pants, and shoes, respectively.
Then the final answer will be
$$A \times B \times C.$$
If Amy and Bunny are selecting $2$ items, from a group of $n$
items, superficially, you would suppose that the number of ways of doing this is $\binom{n}{2}$. However, this overlooks that once the two items (say one green, one red) are selected, then you have to distinguish between [Amy selecting the green item, Bunny selecting the red item] and vice-versa.
Therefore, when computing $A,B,C$, the general formula for each variable will be
$$\binom{n}{2} \times 2.$$
Thus, since the number of shirts, pants and shoes are $5,4,2$ respectively, you have that
$$A = \binom{5}{2} \times 2 = 20,$$
$$B = \binom{4}{2} \times 2 = 12,$$
$$C = \binom{2}{2} \times 2 = 2.$$