Bernhardt (2005) proposed a compensatory model of second language reading. This model predicted that 50% of second language (L2) reading scores are attributed to second language knowledge and first-language (L1) reading ability. In this model, these two factors compensate for deficiencies in each other. Although this model explains a significant portion of L2 reading, much remains unknown. In particular, the compensatory model does not specify the relative contributions of strategic knowledge or background knowledge in L2 reading. Consulting the extant L2 reading literature, this paper proposes a model of second language reading, extending the compensatory model of second language reading. The proposed model predicts the shifting contributions to L2 reading of L2 language knowledge, L1 reading ability, strategic knowledge, and background knowledge. The paper concludes by offering a framework to investigate the explanatory power of the proposed model. (Contains 3 figures.)