Artificial Intelligence has always followed the idea of using computers for the task of modelling human behaviour, with the aim of assisting decision making processes. Scientists and researchers have developed knowledge representations to formalize and organize such human behaviour and knowledge management, allowing for easy translation from the real world, so that the computers can work as if they were “humans”. Some techniques that are common used for modelling real problems are Rough Sets, Fuzzy Logic and Artificial Neural Networks. In this paper we propose a new approach for knowledge representation founded basically on Rough Artificial Neural Networks and Fuzzy Cognitive Maps, improving flexibility in modelling problems where data is characterized by a high degree of vagueness. A case study about modelling Travel Behaviour is analysed and results are assessed.