Note: Just to be clear, I do not refer to "complex molecules" (i.e. "complex" as "complicated"), but to actual complexes.
I am trying to understand the structure of Prussian Blue based on its chemical formula (I ultimately want to explain why it is blue, but this is probably not relevant to this question). This formula is
$$\ce{Fe4[Fe(CN)6]3}$$
where the "first" iron is $\ce{Fe(III)}$ and the "second" is $\ce{Fe(II)}$ but that too is probably not relevant.
Now, having pretty much no experience with complexes, I would read this formula such that for every four $\ce{Fe(III)}$s, there are three $\ce{Fe(CN)6}$s – like this:
(Source)
So far, so good. But I am confused to how this relates to the actual 3-dimensional structure of Prussian Blue which apparently is something like this:
(Source)
I see that from each $\ce{Fe}$, six "bonds" extend to each other $\ce{Fe}$. These "bonds" are, I think, ligands and are actually $\ce{N\equiv C}$. This makes sense because we had $\ce{Fe(CN)6}$ in our original formula. However, there should be some $\ce{Fe(III)}$s, namely four for every three $\ce{Fe-N\equiv C}$. Also, a $\ce{Fe(CN)6}$ shouldn't be able to bond to another because this would mean that there were two $\ce{N\equiv C}$s between each $\ce{Fe}$. This, however, is not the case.
What am I missing here?