In the past, substituting [k_h] for /x/, [g_h] for /G/, and dropping /h/ was common and older loanwords were loaned that way, but in newer loanwords the original Lorošae phonemes are kept. Example: Jinnic "kawro" (originally khawro, Lorošae xawro), but Jinnic fouhız- (Lorošae fouhız-, the dash signifies it's a root and not a word).
By the way, please tell me if I used brackets correctly up there.
The use of brackets is fine, but [g_h] is a rather inexplicable substitution for /G/. In fact, I doubt if it's even phonetically plausible. The "voiced aspirates" of such languages as Hindi and Zulu are more accurately described as being produced with breathy voice
(XSAMPA: [_t]). Moreover, it would be far more bizarre to have a single breathy-voiced segment in a language without this contrast that it would be to have a voiced velar fricative.
IME, speakers of language which lack /G/ tend to hear it as [g]. This is the substitution you find in Arabic borrowings into Swahili, for instance, except voiced stops in Swahili are ingressive (e.g.ǥālin
[Gali] or [g_<ali] "expensive") and it's parallel to your substitution of [k_h] for /x/.