The existing multi-objective optimizers are not so efficient in resolving optimization problems with large number of objectives. For instance, we can mention the multi-objective Grey Wolf Optimizer (MOGWO) used to resolve many engineering problems. Despite its importance, the MOGWO performance is lower, compared to that of other recently-developed optimizers. Thus, this paper presents an Improved Multi-objective Grey Wolf Optimizer (IMOGWO) that allows solving more efficiently multi-objective optimization problems with a large number of objectives while achieving balance between convergence and diversity. The updates in MOGWO enhances the convergence and the exploration of the optimizer. The introduced algorithm was tested on a set of benchmark problems with various numbers of objectives (25 test cases). It was also compared with MOGWO and a set of recent and widely-employed optimizers. The experimental results show that IMOGWO outperformed MOGWO, on all studied benchmark problems, and other comparative algorithms in most test cases. Inferential statistical procedures were applied to highlight the superiority of IMOGWO over other algorithms.