In this study, a modeling approach is developed to simulate the interaction between components in a hybrid-cooled data center. The model considered for this study is a design with a hybrid of air and direct-to-chip liquid cooling. The liquid cooling system comprises pumps, heat exchangers, control valves, quick disconnects, reservoirs, and measuring sensors, all connected in a network of pipes to ensure resilient operation. The right simulation approach enables data center designers to identify key areas of efficiency improvement during the dynamic operation of the system in advance. This will save a significant amount of time and cost while allowing the assessment of various cooling solutions. Precise modeling provides valuable insights which can prevent catastrophic failures in data centers. The current study deploys Computational Fluid Dynamics (CFD) together with 1D Flow Network Modeling (FNM) to offer high-accuracy prediction within a short time of computation. Equipment sizing, redundancy analysis, air/liquid flow distribution, flow resistance characteristics, and temperature changes in both transient and steady-state conditions are direct outputs of a combined CFD and FNM. The CFD provides the capability to evaluate the effectiveness of air cooling within the racks and aisles. Furthermore, it helps to estimate air streamlines in an airflow network of a data center looking for possible recirculation and room pressure imbalances. The proposed approach enables the execution of extensive concept studies and modeling of complex hybrid cooling systems at any data center scale with the required level of resolution. In a gradual approach, modeling starts from the component level moving to the server, then the rack, and later reaches the data center level. Key inputs for both CFD and FNM are discussed in detail and a comprehensive example is presented to navigate through each step. It is of vital importance to use engineering judgment to interpret results confirming physics is followed correctly.