Optimization of the association between wireless stations and access points (APs) has shown its effectiveness to improve the overall performance of wireless LAN. Most of the previous works do not consider the latest amendments of the IEEE 802.11 standard. The main challenges are to propose models that take into account recent enhancements such as spatial multiplexing (MIMO) at the physical layer and frame aggregation mechanism at the MAC layer. To assess these new features, we derive an association optimization approach based on a new metric, named Hypothetical Busy Time Fraction (H-BTF), that combines the classical Busy Time Fraction (BTF) and the frame aggregation mechanism. This metric is based on local measurements like throughput demand and frame error rate for each station. The model estimates the H-BTF of each AP for any configuration and is thus able to predict H-BTF for other association scheme. Association is then optimized to minimize the load of the busiest APs. This load balancing between APs aims to satisfy stations with regard to their throughput demands. Numerical evaluations performed with the network simulator ns-3 have shown the accuracy of the proposed approach for a large set of scenarios and a significant benefit for the stations in terms of throughput and satisfaction.