Limiting low- and high-pressure rate coefficients as well as full falloff curves have been modeled by unimolecular rate theory for the recombination reaction 2CF(3) (+ M) -> C2F6 (+ M) and the reverse dissociation of C2F6. The results are compared with experimental data from the literature. Although there are considerable discrepancies (up to a factor of 5) between various experimental data near 300 K and the database for high temperatures is still limited, we try to conclude on the temperature dependence of the high-pressure rate coefficient. We suggest that there is only a small and probably positive temperature coefficient of the latter quantity. The present theoretical modeling seems to be in agreement with this experimental result, but it is in disagreement with conclusions from earlier theoretical work. The difference is attributed to different empirical assumptions about the anisotropy of the potential. It is shown that nearly all previous experiments (except high-temperature shock wave and very low pressure pyrolysis/photolysis experiments) correspond to nearly limiting high-pressure conditions.