This paper presents the development of an improved spiral dynamic optimization algorithm with application to nonparametric fuzzy logic modeling of a twin rotor system (TRS). The spiral motion and dynamic step size are generated by a spiral model that produces unique exploration and exploitation strategies of a spiral dynamic algorithm (SDA). However, the algorithm is subjected to settling into local optima at the end of the search process, due to insufficient exploration throughout the search area. An elimination-dispersal strategy of a bacterial foraging algorithm is adopted to solve the problem. The proposed strategy is validated with various benchmark functions and a nonparametric test analysis is performed to evaluate its significant achievement in comparison to the SDA. Moreover, nonlinear dynamic modeling of a TRS in hovering mode is presented and the results show the effectiveness of the proposed algorithm to solve real-world problems. The result of the modeling work is presented in both time-domain and frequency-domain. It shows how the fuzzy model optimization represents the characteristic behavior of the TRS very well.