Purpose: This study was conducted to investigate the effect of principal leadership (PL) on teacher professional learning (TPL) with the possible mediation of teacher trust (TT) and teacher agency (TA) using a broader set of data collected from published research. Design/methodology/approach: The study uses meta-analytic structural equation modelling to compare weighted effect sizes between variables produced by research across different periods and contexts. The bilateral relations of PL with study variables were first examined, and then the model-data fit was tested for both the proposed and the two alternative models. Findings: The results showed that PL had a significant and "very large" effect on TPL, TT and agency; PL affected TPL mostly indirectly over TT and agency. TT and agency collectively predicted nearly half of the variance in TPL. Originality/value: The study uses a novel and advanced methodology to investigate the relationships among the study variables using data from 59 studies with a sample size of 67,035 participants from around the world. The results contribute to the literature, offering a more global understanding of the interaction of PL, TPL, trust and agency.