<p>(Yousefi Heris, Erkenntnis, 2024) offers a version of simulation theory, building on (Goldman, Simulating minds: The philosophy, psychology, and neuroscience of mindreading, Oxford University Press, 2006) and (Herschbach, Synthese 189:483–513, 2012), that refines how to determine the target and content of a simulation. In this critical discussion I will raise some challenges for his account. I first question whether he offers a genuine advance over Herschbach’s account of how to determine the target of a simulation. Then I focus on his account of how to ground the content of simulation. Yousefi Heris surveys three options: the Neural Correspondence Thesis (NCT), the Functional Correspondence Thesis (FCT), and the Context-Dependent Correspondence Thesis (CDCT). He believes only the CDCT can accommodate a modern picture of our brain’s architecture. I disagree with Yousefi Heris’s position that the CDCT is needed to fit with an anti-localizationist picture of the relations between neural states and mental states. Since defending that position would require a paper of its own, I focus on exposing an unacknowledged implication of Yousefi Heris’s account: that it abandons ST’s notions of replication simulation, making it better characterized as a form of the theory theory.</p>