Understanding how humans move is a key factor for the design and evaluation of networking protocols and mobility management solutions in mobile networks. This is particularly true for mobile scenarios in which conventional single-hop access to the infrastructure is not always possible, and multi-hop wireless forwarding is a must. We specifically focus on one of the most recent mobile networking paradigms, i.e., opportunistic networks. In this paradigm the communication takes place directly between the personal devices (e.g., smartphones and PDAs) that the users carry with them during their daily activities, without any assumption about pre-existing infrastructures. Among all mobility characteristics that may affect the performance of opportunistic networks, the users' travelling patterns have recently gained a lot of attention due to their impact on the spreading of both viruses and messages in such a network. In this paper we consider a social-based mobility model (HCMM) and we extend this model to account for the typical travelling behaviour of users. To the best of our knowledge, the resulting mobility model is the first model in which movements driven by social relations also match statistical features of travelling patterns as measured in reality. Finally, we evaluate our proposal through simulations over a wide range of scenarios, emphasizing the effect of finite sampling on the obtained results.