In this paper, we define and study convergence of an Ishikawa type iteration scheme with a new and weaker control condition for two nonself asymptotically quasi-nonexpansive mappings on a uniformly convex Banach space. Our results improve and extend the corresponding results of Khan, Takahashi [4], Rhoades ]6], Tan, Xu [9], Chidume, Ofoedu, Zegeye [14], Wang [15], and Qihou [18].