In the previous work of Huang et al., a decentralized adaptive fuzzy controller of large-scale multiple-input multiple-output nonlinear systems is obtained predicated upon the assumption that the interconnections between subsystems can be bounded by first-order polynomials. In this note, we focus in the absence of the conservative assumption upon developing a novel decentralized adaptive fuzzy control scheme. In virtue of fuzzy systems and a regularized inverse matrix, the developed control scheme not only addresses controller singularity under an overall design framework but also copes with interconnections with arbitrary nonlinear bounds. The resulting closed-loop large-scale system is proved to be asymptotically stable. The controller design is applicable to an automated highway system, and simulation results confirm its practical usefulness.