To study the safety and stability of the main truss arch structure during the concrete pouring process in a long-span CFST arch bridge, a combination of theoretical analysis and numerical simulation was adopted. The world's longest span CFST arch bridge, the Pingnan Third Bridge (main span 575 m) in Guangxi Province, was taken as an example. Firstly, the basic theory of stability and the concrete pouring technology of the main truss arch structure were summarized. Secondly, a three-dimensional finite element model of the main truss arch structure was established for the concrete pouring stage. Finally, the linearity, stress, and stability of the main truss arch structure during concrete pouring were studied. The results show that both vertical and transverse deflections occurred in the main truss arch structure during concrete pouring from the arch foot to the arch vault; the upper and transverse deflections of the main truss arch structure showed parabolic changes, and the upper and transverse deflections of the midspan section were larger than those of other sections. When concrete was poured near the 1/6 section of the main truss arch structure, the upper and transverse deflections of the middle section of the main truss arch structure peaked; the maximum upper deflection reached 113 mm, and the maximum transverse deflection reached 60 mm. In the process of concrete pouring from the arch foot to the arch vault, the stress in the steel tubes and concrete was constantly changing. Although upper and transverse deflections of the main truss arch structure occurred during concrete pouring, the stress in the steel tube and concrete in the main truss arch structure did not exceed the tensile strength, and the linearity of the main truss arch could be adjusted by retaining the anchor cables, which were set in the erection stage of the main truss arch structure at the 1/6 section of the main truss arch structure. During the concrete pouring process of the various steel tubes of the main truss arch structure, the stability coefficients of the main truss arch structures decreased gradually. During concrete pouring in a single steel tube, the maximum reduction in the stability factor was 21.32%, which occurred in the 1# steel tube, and the minimum decrease in the stability factor was 8.89%, which occurred in the 8# steel tube. The stability coefficients were greater than 4 after this reduction and met the requirements of the design specifications. This study can provide a reference for the optimization design and construction monitoring of long-span CFST arch bridges.