This study aims to produce a holistic-integrative PAI learning model in the cognitive, affective, and psychomotor domains in an integrated manner (PHIKAP) using Capra's theory which has the principles of connectedness, wholeness, and being. This research uses a Research and Development (R & D) approach based on the theory from Borg & Gall, which is simplified into four stages: a preliminary study, designing a model draft, validating the model draft; and testing the model in the field using a quasi-experimental Pretest and Posttest Non-equivalent Control Group Design. Collecting data using questionnaires, interviews, trials, and tests. Data analysis used quantitative and qualitative methods (mixed method). For quantitative data using the Mann-Whitney U test statistic and qualitative data using descriptive qualitative. Research results: (1) PAI learning in elementary schools, which is currently taking place, is more dominant in the cognitive domain than affective and psychomotor, even though these three domains should ideally be carried out in a holistic, integrative manner according to juridical, psychological, and religious provisions; (2) The product of this research is the PHIKAP PAI Model in Elementary Schools, which is a learning design that is structured starting from planning, implementing, and evaluating connectedness, wholeness and being in the cognitive, affective and psychomotor domains. The model developed includes rationale, goals, principles, planning, implementation, and evaluation strategies, as well as criteria and assumptions for the success of Islamic education learning in elementary schools; (3) The effect of the PHIKAP model is proven to be effective in improving the quality of the learning process by 75.75% through strengthening the functions of the cognitive, affective, and psychomotor domains in a comprehensive and balanced manner. This value is higher than the group that does not use the PHIKAP model with a quality value of 58,63..%; (4) The effect of the PHIKAP model is proven to be effective in improving the quality of PAI learning outcomes in the cognitive domain by 85.97%, 91.43% in the affective domain, and by 89.18% in the psychomotor domain. This value is higher than the group that does not use the PHIKAP model.