help! phonology...

I'm working my way through an introductory phonology/phonetics book and
trying to overhaul the phonology of Chevraqis.  What I have, tentatively,
is (using Kirschenbaum):

Vowels:
i u
E O
 a

Consonants:
(p) b   t d        c   k
 m      n
        *
(f) v   s z  S Z   C   x

Clusters can only be of the form consonant-plus-*, or /tS/ or /dZ/.
[b] manifests as /p/ in word-final position, /b/ elsewhere.
[v] manifests as /f/ in word-final position, /v/ elsewhere.
(I *think* this rule makes some amount of sense but I'm not entirely
confident.)

(Having heard /c/ I have become rather enamored of the sound).

I've honestly been worrying about this whole symmetry of sound systems
business.  I keep staring at the IPA chart and I can't figure out how to
justify the palato-alveolars /S/ and /Z/.

Also, (p) b, t d, k
is pretty unsymmetric, but I really, *really* dislike /g/ aesthetically
and have been running around in circles trying to figure out if it makes
sense to not include /g/.  How strict/common is this symmetry principle?
I'm almost prepared to lose all the voiced versions of sounds, but I
wanted to keep /t/ and /d/, /s/ and /z/, and /S/ and /Z/ for contrasted
but easy-to-remember inflections for dynamic-vs.-static conjugations (the
only thing changing in the inflection would be the voicing).  Can I get
away with this?

A question on transcription, for those who've made it so far:  I'm
contemplating using these Romanizations:

tj for /c/
sj for /C/
sh for /S/
zh for /Z/

I think the latter two are fairly "easy" for an English-speaker to figure
out, but I'm not really sure what to do with /c/ and /C/, especially
since I'm using "ch" for /tS/.

Help?

YHL