One of the largest sources of data on radiation exposure in humans is the study of the atomic bomb survivors at Hiroshima and Nagasaki, Japan performed by the Radiation Effects Research Foundation (RERF). As part of their retrospective dosimetry efforts for the atomic bomb survivors, RERF published two core systems: Dosimetry System 1986 (DS86) and Dosimetry System 2002 (DS02). Due to computing limitations at the time, only three stylized phantoms (an infant, child and adult) were used in DS86 and DS02 to represent the entire Japanese population. In this study, we sought to evaluate the dosimetric differences that should be expected from using an updated and age-expanded phantom series with the survivor cohort. To this end, we developed a new series of hybrid phantoms, based on the Japanese population of 1945, which has greater anatomical realism and improved age resolution over those used by RERF. These phantoms were converted to voxel format and compared to their older counterparts through the calculation of organ dose coefficients using DS02 free-in-air particle fluences at three distances from the bomb hypocenter. From the photon portion of the spectra, organ dose differences of up to nearly 25% are expected between the old and new series, while organ dose differences of up to nearly 70% are expected from the neutron portion. We also compared organ dose coefficients among themselves to determine the accuracy in the use of one organ dose as the epidemiological surrogate to another. Certain organ-surrogate pairs were shown to be inappropriate, such as the use of colon dose for breast risk analyses. Overall, our new series of phantoms provides significant improvements to survivor organ dosimetry, especially to those survivors who were previously misrepresented in body size by their stylized phantom and to those who experienced a highly-directional irradiation field.