The main concern of clustering approaches for mobile wireless sensor networks (WSNs) is to prolong the battery life of the individual sensors and the network lifetime. In this paper, we propose a balanced and safe weighted clustering algorithm which is an extended version of our previous algorithm (ES-WCA) for mobile WSNs using a combination of five metrics. Among these metrics lie the behavioral level metric which promotes a safe choice of a cluster head in the sense where this last one will never be a malicious node. The goals of the proposed algorithm are: offer better performance in terms of the number of re-affiliations which enables to generate a reduced number of balanced and homogeneous clusters, this algorithm, coupled with suitable routing protocols, aims to maintain stable clustering structure. We implemented and tested a simulation of the proposed algorithm to demonstrate its performance.