In power systems with high proportion of variable renewable energy, the scarcity of inertia and primary frequency response (IPFR) becomes a critical issue. This evolution necessitates the emergence of corresponding markets. The increasing variety of markets and the diversity of market participants have led to more complex bidding behaviors than before, which need be thoroughly studied. This paper proposes a bi-level model to analyze the bidding behaviors of a renewable-storage system (RSS) acting as a price maker in multiple markets. The nonlinear relationship between IPFR and system frequency is modeled. To depict the characteristics of IPFR and future markets, the unit commitment (UC) process is embedded. To address the nonconvexity caused by the UC process in the proposed bi-level model, a solution approach based on penalty function and dual theory is presented. The proposed model and its solution method are applied to a case study based on the IEEE30-bus system and historical operational data from the California Independent System Operator. The case study results illustrate that the proposed model can effectively characterize the complex bidding behaviors of RSS in multiple markets and validate the efficacy of the solution method.