Apatite (U-Th)/He (AHe) thermochronology depends on accurate knowledge of how diffusion occurs. This involves measurement of core diffusion kinetics as well as understanding the behavior of migrating He atoms. Drawing from previous studies as well as data obtained via continuous ramped heating (CRH), we assess several processes that need to be integrated into a single model for He diffusion in apatite. CRH analyses conducted at different heating rates show a kinetic response for both the “normal” lower-temperature and the higher-temperature release peaks, with peaks shifting to lower temperatures at lower heating rates. Where we do see a rollover in Arrhenius trends it also shows a kinetic response, being deferred to higher temperatures at higher heating rates, though many samples with unimodal release peaks do not show a significant rollover; fluorapatites seem to show more prominent rollover. For samples showing multiple release peaks, we find that their Arrhenius data often transition from one lower-temperature trend to another at higher temperatures that has about the same slope and thus activation energy. This looks very much like MDD behavior in K-feldspar, and MDD domain analysis fits the observed data very well, even if mechanisms involving discrete domain sizes are implausible. This interesting and unexplained result must speak to the nature of what is happening during analysis of samples having trapped He. To explore our data, we coded a simple diffusion model in which single He atoms are free to jump within a grid, but can also arrive at grid nodes designated as reversible sinks, escape from which depends on an exponentially temperature- dependent probability. The model includes radiogenic He production over geological thermal histories followed by laboratory CRH outgassing. When conditioned using D values observed for AHe, the model accurately predicts parameters such as closure temperature and fractional loss. When traps are introduced, the model simulates the essential nature of the dual-peak CRH results we see. Three important results emerge from this model. (1) Few sinks need be present. (2) Trapping occurs twice during diffusion, first in nature and then again during laboratory outgassing, meaning that the ratio of the gas amounts beneath each CRH peak overestimates the geological trapping. (2) Trapping in nature is very dependent on the sample’s thermal history: it is smallest for ancient rapid cooling and largest for samples that reside in the PRZ (allowing more radiogenic production to find traps before diffusion ceases). This model raises the possibility that complex CRH data record extended thermal-history information. If CRH and 4He/3He analysis were combined the 3He lab outgassing would record the sample’s trapping dynamics, and the 4He outgassing would reflect that plus a segment of the sample’s thermal history, which could be extracted using the 3He observations.