An important topic in the theory of state-to-state chemical reactions is the way in which variations in the potential energy surface (PES) can affect product differential cross sections (DCSs). We investigate this topic for three canonical state-to-state reactions at a translational energy of 62.1 meV, namely: F + H2(vi = 0, ji = 0, mi = 0) → FH(vf = 3, jf = 0, 1, 2, mf = 0) + H, where vi, ji, mi and vf, jf, mf are the initial and final vibrational, rotational and helicity quantum numbers respectively. We use the state-of-the-art 2015 ab initio ground-state PES of Chen-Sun-Zhang (CSZ). We compare our results with those we obtained earlier for the 2008 PES of Fu-Xu-Zhang (FXZ), which were reported in a companion paper by Xiahou, Connor, De Fazio, Sokolovski [Phys. Chem. Chem. Phys., 2024, 26, 3647]. Emphasis is placed on the continuation of the discrete scattering matrix to continuous complex values of the total angular momentum and the resulting properties of the Regge pole positions, residues and life-angles, as these topics are little understood. In general, we find the CSZ and FXZ results change in a robust way. This is encouraging because it implies further refinement of the ground-state CSZ PES will make it sufficiently accurate for use in a wide range of dynamical computations. This paper contains the following topics: (i) results for accurate quantum scattering matrix elements for the CSZ PES for indistinguishable H2. (ii) Full and nearside-farside (NF) DCSs and local angular momenta, including two resummations of the partial wave series for the full scattering amplitude. (iii) The "CoroGlo" test is applied, which suggests that the forward-angle scattering is a glory and not a corona. (iv) Six asymptotic theories are used to prove there is a glory at forward angles. (v) Three asymptotic theories prove there is a broad (hidden) farside rainbow at sideward-scattering angles. (vi) We calculate Regge pole positions and residues using Thiele rational interpolation in the complex angular momentum (CAM) plane. At the translational energy of 62.1 meV, there is just one numerically stable pole in the first quadrant of the CAM plane. (vii) Using Regge theory, we find that the single pole contributes to the DCS across the whole angular range. This proves, in a mathematically sound and unambiguous manner, that a resonance Regge pole contributes to the DCS for all three transitions, being most prominent at forward angles. (viii) The diffraction oscillations in the DCSs arise from NF interference. Also, in the CAM theory there is interference between the resonance Regge pole subamplitude and the direct subamplitude. (ix) Our emphasis is on quantitative theories of reactive scattering and the avoidance of subjective qualitative interpretations.