Single-cell sequencing has dramatically enhanced the scale and resolution of molecular profiling for tissues and organs. We have recently generated an integrated dual-modal reference atlas of the retina, the most accessible part of the mammalian central nervous system. We compiled approximately 3.9 million cells from 125 donors, incorporating data from eight published studies and 2.7 million unpublished data points, to create a comprehensive Human Retina Cell Atlas (HRCA) identifying over 120 cell types. Our analysis revealed intriguing differences in transcriptomes, chromatin landscapes, and gene regulatory networks (GRNs) across various cell types. Additionally, we modeled changes in gene expression and chromatin accessibility across age, ancestry, and tissue regions. This integrated atlas facilitated the fine-mapping of genome-wide association study (GWAS) and expression quantitative trait loci (eQTL) variants. Further improvements to the Version 1 HRCA are planned, incorporating single-cell spatial transcriptomics and single-cell long-read RNA sequencing, which will enhance our understanding of retinal function and pathology.