In this work, we introduce a scheme for modelling actor systems within sequential term rewriting. In our proposal, a TRS consists of the union of three components: the functional part (which is specific of a system), a set of rules for reducing concurrent actions, and a set of rules for defining a particular scheduling policy. A key ingredient of our approach is that concurrent systems are modelled by terms in which concurrent actions can never occur inside user-defined function calls. This assumption greatly simplifies the definition of the semantics for concurrent actions, since no term traversal will be needed. We prove that these systems are well defined in the sense that concurrent actions can always be reduced. Our approach can be used as a basis for modelling actor-based concurrent programs, which can then be analyzed using existing techniques for term rewrite systems.