When we get distracted, we may engage in mind-wandering, or task-unrelated thinking, which impairs performance on cognitive tasks. Yet, we do not have cognitive models that make this process explicit. On the basis of both recent experiments that have started to investigate mind-wandering and introspective knowledge from for example meditators, we built a process model of distraction in the form of mind-wandering. We then tested the model by predicting performance on tasks used in mind-wandering studies. We showed that we could both predict task performance as well as the participants' responses to questions about what they were thinking about. This improved understanding of mind-wandering could be used in the future to revise our models of when, how, and why distraction occurs. For example, our model could be used to examine how the effect of distraction on task performance depends on the type of mind-wandering (e.g., rumination versus day-dreaming).