The inclusion of the practice of "developing and using models" in the "Framework for K-12 Science Education" and in the "Next Generation Science Standards" provides an opportunity for educators to examine the role this practice plays in science and how it can be leveraged in a science classroom. Drawing on conceptions of models in the philosophy of science, we bring forward an agent-based account of models and discuss the implications of this view for enacting modeling in science classrooms. Models, according to this account, can only be understood with respect to the aims and intentions of a cognitive agent (models for), not solely in terms of how they represent phenomena in the world (models of). We present this contrast as a heuristic--"models of" versus "models for"--that can be used to help educators notice and interpret how models are positioned in standards, curriculum, and classrooms.