In the last years, multi-objective genetic algorithms have been successfully applied to obtain Fuzzy Rule-Based Systems satisfying different objectives, usually different performance measures. Recently, multi-objective genetic algorithms have been also applied to improve the difficult trade-off between interpretability and accuracy of Fuzzy Rule-Based Systems, obtaining linguistic models not only accurate but also interpretable. It is know that both requirements are usually contradictory, however, a multi-objective genetic algorithm can obtain a set of solutions with different degrees of accuracy and interpretability. This contribution briefly reviews the state of the art in this very recent topic and presents an approach in order to prove the ability of multi-objective genetic algorithms for getting compact fuzzy rule-based systems under rule selection and parameter tuning, i.e., to obtain linguistic models with improved accuracy and the least number of possible rules. This way to work involves another trade-off degree respect with the works in the existing literature which has been still not explored.