Proto-Muskogean *N and Timucua t?

Some Proto-Muskogean correspondence sets show a correspondence between /n/ in Western Muskogean and /ł/ in other Muskogean languages.  An example is the following set for FISH:

Choctaw and Chickasaw  nani'
Alabama & Koasati          łało
Hitchiti & Mikasuki         ła:ł-i
Creek                                łało

Mary Haas reconstructed this phoneme as *N, but its phonetic realization has always been somewhat uncertain.

I have begun to notice some cases where this PM phoneme matches with /t/ in Timucua:

FATHER   PM *iNki   Timucua iti
BROTHER (of a woman)  PM *NakfV   Timucua tafi 'brother-in-law (of a woman)'
ARROW PM *Naki  Timucua atulu?

Several PM kin terms end in -ki, which is probably separable.  If the BROTHER word contains a metathesized version of the same -ki, then possibly the pre-Proto-Muskogean are

FATHER pPm *iN(i)
BROTHER pPM *Naf(i)

This relationship might make sense if PM*N = θ.