Rotary near-field lithography (RNFL) technology provides a route to overcome the diffraction limit with a high throughput and low cost for nanomanufacturing. Utilizing the advantage of the passive flying of a plasmonic head, RNFL can achieve a 10 m/s processing speed with a perfect near-field condition at dozens of nanometers. The flying performance of the plasmonic flying head (PFH) is the pivotal issue in the system. The linewidth has a strong correlation with the near-field gap, and the manufacturing uniformity is directly influenced by the dynamic performance. A more serious issue is that the unexpected contact between the PFH and substrate will result in system failure. Therefore, it is important to model and analyze the flying process of the PFH at the system level. In this study, a novel full-coupled suspension-PFH-air-substrate (SPAS) model that integrates a six-degree of freedom suspension-PFH dynamics, PFH-air-substrate air bearing lubrication, and substrate vibration, is established. The pressure distribution of the air bearing is governed by the molecular gas lubrication equation that is solved by the finite element method (FEM) with a local pressure gradient based adaptive mesh refinement algorithm using the COMSOL Multiphysics software. Based on this model, three designs of the air bearing surface are chosen to study the static, dynamic, and load/unload performance to verify whether it satisfies the design requirements of RNFL. Finally, a PFH analysis solver SKLY.app is developed based on the proposed model.