The contradiction between economic development and ecological protection has become a common challenge for territorial governance in developing countries around the world. However, extant studies have neglected the coupling and symbiotic relationship between humans and nature, resulting in significant functional conflicts, insufficient stability, and imbalances in ecological and economic benefits in the reconstruction of territorial spatial functional pattern (TSFP), making it difficult to achieve synergies between development and protection. The question that arises is how the TSFP can be reconstructed in order to achieve harmonious coexistence between humans and nature. This remains a challenging problem in the context of the synergizing development and protection of the TSFP. This study innovatively integrates set-theoretic principles and functional mappings with game-theoretic analysis to develop Territorial Spatial Functional Pattern Reconstruction (TSFPR) model designed to foster harmonious human–nature coexistence, and validates the model using geospatial data from Qionglai City, China. Empirical evidence demonstrates that, in comparison with conventional methods, TSFPR model significantly mitigates the territorial spatial functional conflicts (TSFCs), enhances stability and ecological and economic benefits, and achieves the expected harmonious coexistence between humans and nature. The analysis confirms that the territorial spatial functional conflict (TSFC) coordination index established in this study provides a reliable criterion for identifying superior territorial spatial functions (TSFs). The proposed TSFPR model is an expansion of the theory of spatial optimization modelling, and it provides a tool for reconstructing the TSFP for the harmonious coexistence between humans and nature. In summary, the utilization of the TSFPR model to reconstruct the TSFP for harmonious coexistence between humans and nature provides a novel solution for coordinating the development and protection of territorial space governance.