In this paper, it is shown that stable model semantics, perfect model semantics, and partial stable model semantics of disjunctive logic programs have the same expressive power with respect to the polynomial-time model-equivalent reduction. That is, taking perfect model semantics and stable model semantic as an example, any logic program P can be transformed in polynomial time to another logic program p such that perfect models (resp. stable models) of P 1-1 correspond to stable models (resp. perfect models) of P, and the correspondence can be computed also in polynomial time. However, the minimal model semantics has weaker expressiveness than other mentioned semantics, otherwise, the polynomial hierarchy would collapse to NP.