Constructing explanations can have instrumental and intellectual benefits. The instrumental benefits are cognitive states that are produced by explanations and that, together with other factors, determine somehow how we will act in the future. The intellectual benefits are states that are produced by explanations and that have an intrinsic positive value, i.e., states of mind that we like in se, and not for an external reason. Like many other types of explanations, so-called "folk-psychologica\" explanations of human actions (explanations in terms of desires, beliefs, norms, etc.) have certain instrumental benefits. These benefits can be divided into two categories. On the one hand, such explanations can be used to determine our legal, moral and sentimental attitude towards an action. Should the agent be punished? Was his/her action morally right? Should we be grateful, forgive, be angry, ¼? Our answers to these questions depend on the causes of the actions as described in the explanation. On the other hand, explaining a person's action might help us to control his/her future actions in similar circumstances. The principal aim of this article is to answer the following question: what format should explanations of actions have to ensure that they have the second type of instrumental benefit? Most of section 2 will be devoted to this question. This does not mean that we disregard the instrumental benefits of the first type: at the end of section 2, we answer the analogous question for the first type. In section 3 we embed our approach to the explanation of action into a general theory of explanation. We do this by opposing our approach with Carl Hempel's deductive-nomological model, Peter Lipton's contrastive approach, Paul Humphreys' aleatory model and Wesley Salmon's causal-mechanical model. In section 4, we compare our model with other models for the explanation of action: Carl Hempel (once more), Georg H. von Wright and Raimo Tuomela.