Prompt gamma-ray activation analysis has long been used for nondestructive elemental analysis, particularly of the light elements (e.g. H, C, N, S, P, Cl). However, the accuracy and precision of the method for elemental determination of hydrogenous materials is limited by the calibration of elemental sensitivities (cps/mg element), which vary with both neutron scattering power (i.e. hydrogen content) and target geometry. In addition, sensitivities for many of the low Z elements (C, N, S, P) are poor, yielding poor counting statistics without extremely long irradiations. We address these issues with these three approaches: 1) A combination of cold and thermal neutron PGAA continues to be used to improve both accuracy and precision of light element measurement in hydrogenous materials. CNPGAA yields higher signals due to the greater neutron capture cross sections for cold neutrons, but the absolute sensitivity varies as sample mass, a complication of a greater neutron scattering cross section. The latter issue is addressed by performing an absolute sensitivity calibration for selected matrix elements at the TNPGAA, where the scattering effect is minimal. This way, we take full advantage of the CNPGAA for efficient and precise determination of the sensitivity RATIOS, and improve the accuracy of the absolute elemental concentration via sensitivity calibration at the TNPGAA. The TNPGAA calibration essentially serves as a correction for scattering effects in CNPGAA. We have previously reported use of this method for determination of sulfur in fuel oil reference materials. More recent measurements include the determination of carbon at 12.5 % mass fraction in dolomite limestone reference material with an expanded uncertainty of 1.3 %, and C, N, S, and other elements in coal SRMs. Using both methods combined avoids the possible analytical bias from neutron scattering in CNPGAA and the poorer uncertainties from counting statistics from TNPAA. 2) The calculation of element sensitivities using tabulated k0 factors or partial capture cross sections and carefully measured detector efficiencies also serves as a useful check on elemental sensitivities measured from standards. CNPGAA measurements of C, H, and N in plastics using cross sections agreed with measurements made using comparator standards to within a few percent, indicating that the cross-section method may be used to verify measurements made using the standards comparison method, or for standard-free PGAA measurements when high precision is not needed. 3) The determination of gamma-ray background can also be a hindrance to detection limits, especially in the measurement of hydrogen. Measurements of ultra-trace amounts of hydrogen in materials require careful measurement of H in packaging materials as well as investigation of possible low-level interference peaks from other methods. Methods for studying these interferences are ongoing.