Recently, operators have resorted to femtocell networks in order to enhance indoor coverage and increase system capacity. Nevertheless, to successfully deploy such solution, efficient resource allocation algorithms and interference mitigation techniques should be deployed. The new applications delivered by operators require large amounts of network bandwidth. Whereas, some customers may want to pay more in exchange for a better quality of service (QoS), some others need less resources and can be charged accordingly. Hence, we consider an OFDMA femtocell network serving both QoS constrained high-priority (HP) and best-effort (BE) users. Our objective is to satisfy a maximum number of HP users while serving BE users as well as possible. This optimization problem is multi-objective NP-hard. For this aim, we propose in this paper a new resource allocation and admission control algorithm, called Q-FCRA, based on clustering and taking into account QoS requirements. We show by extensive network simulation results that our proposal outperforms two state of the art schemes (Centralized-Dynamic Frequency Planning, C-DFP, and Distributed Random Access, DRA) as well as our previous proposal, FCRA, in both low and high density networks. The results concern the number of accepted users, the fairness, the throughput satisfaction rate and the spectrum spatial reuse.