Today I came across the following theorem of Mersmann: There are precisely 14 primitive eta-products which are holomorphic modular forms of weight $\frac{1}{2}$, namely...
He also proves a conjecture of Zagier to the effect that there are essentially only finitely many such products of any given weight.
I learnt these facts from Zagier's contribution to the The 1-2-3 of modular forms (which happens to have been metioned by Wadim Zudilin).
Addendum (2011/02/10) Came across an advertisement for the book Eta products and theratheta series identities by Günter Köhler.