Document reranking (DR) and next query prediction (NQP) are two core tasks in session search, often driven by the same search intent. So far, most proposed models for DR and NQP only focus on users’ short-term intents in the current search sessions. This limitation fails to recognize and address the long-term intents present in historical search sessions. We consider a personalized mechanism for learning a user’s profile from their long-term behavior to simultaneously enhance the performance of DR and NQP in an ongoing session. We propose a personalized session search model, called Long short-term session search Network (LostNet), that jointly learns to rerank documents for the current query and predict the next query. LostNet consists of three modules: a hierarchical session-based attention mechanism, a personalized multi-hop memory network, and a DR and NQP network. The hierarchical session-based attention mechanism tracks fine-grained short-term intent from a user’s search session. The personalized multi-hop memory network tracks a user’s dynamic profile information from their search sessions so as to infer their personal search intent. The DR and NQP network reranks documents and predicts the next query synchronously based on outputs from the above two modules. We conduct experiments on two benchmark session search datasets. The results show that LostNet achieves significant improvements over state-of-the-art baselines.