A major limitation to such combinations is the accumulation of side-effects of individual drugs which constitutes a major burden on the human physiology, and may lead to non-compliance.

Interesting thought, however,it will be good to study this in an animal model of schizophrenia. In my opinion, you are more likely to create a 'zombie' rather than a fully-functioning human, thereby defeating the main purpose of therapy.

SSRIs, like MAOIs, are very dangerous in combination with antipsychotics because of the high risk of resulting hyperserotonemia (serotonin syndrome), potentially being a fatal syndrome.  A psychostimulant on top of that could produce even more drastic outcomes.

Atypical antipsychotics are serotoninergic ANTAGONISTS whereas SSRI are indirect agonists - therefore no risk for serotonin syndrome is present. Moreover, modafinil is not the ordinary psychostimulant as it has central orexin/histamine actions that would counteract some of the side effects of both olanzapine and fluoxetine, and a mood stabilizer would keep the drugs 'cerebral tension' in control. The combination was selected carefully and requires a carefully selected answer. 

Point well made about the contrasting drug types, but antagonists don't necessarily stop the flooding of endogenous serotonin, dopamine, epinephrine, etc -- they simply have better binding affinity.  Many people suffering from psychosis have a flood of dopamine occurring as it is, and, if at all manic, serotonin as well.  An SSRI would, in my opinion, be a questionable choice there by, not exactly due to any psychological effects but to physiological ones (including neuroleptic malignant syndrome).   What gives you the inclination towards such a proposal with SSRI use?  Don't be too hasty in assuming that I am not being careful with my responses.  It's unprofessional.  Discretion advised.

Schizophrenia is part of the psychotic spectrum and the combination of fluoxetine and olanzapine particularly is considered as the best one for bipolar depression, thus why not adopt it to schizophrenia. In apsychotic schizophrenic patients pseudonegative/negative/cognitive symptoms/deficits are observed, most probably due to a hypofrontality, particularly hypofunction of the dopaminergic D1 receptors in the prefrontal cortex (DLPFC). Therefore, a potentiation of the dopaminergic neurotransmission in the prefrontal cortex (where D1 receptors are abundant) would ameliorate the abovementioned negative sides of schizophrenia, considered as dementia praecox and not the acute psychotic state. Both olanzapine and fluoxetine increase dopamine and noradrenaline levels in PFC via 5HT2C antagonism, whereas fluoxetine, modafinil and valproate activate agonistically 5HT1A receptors in the PFC, which have a potentiating effect on the dopaminergic neurotransmission in PFC. Olanzapine is a histamine H1 antagonist, and modafinil would counteract this unnecessary effect, and, moreover, will potentiate dopaminergic, noradrenergic, serotoninergic and histaminergic neurotransmissions in PFC. Olanzapine has no pro-serotoninergic actions, unlike clozapine. Valproate will counteract the elevated glutamate levels in PFC, possibly protect against a re-emergence of psychosis and increase signal-to-noise ratio in PFC info flow (note that NMDA receptors are autoreceptors as well and in acute psychosis we have overactivation of other and postsynaptic NMDA receptors due to this fact). The well-known fact that PFC has the closest connections with the formatio reticularis and exerts regulatory control over it in our case would translate to a return of the control to this higher structure, which was lost due to the neurodegeneratory events in PFC particularly, and ultimately restore enough power to the patient to find a comfortable place in the intersubjective world to live in. There are some pharmacokinetic interactions that might arise, but this can be controlled. Animal studies could be helpful. A risk of a serotonin syndrome might arise, in my opinion, only due to the serotoninergic actions of modafinil but I suspect that they are not so pronounced (have to check).      

Concerning the dubious role of a SSRI augmentation check this: Br J Psychiatry. 2010 Sep;197(3):174-9. Efficacy of antidepressants in treating the negative symptoms of chronic schizophrenia: meta-analysis. Singh SP1, Singh V, Kar N, Chan K.

Conclusions

Antidepressants along with antipsychotics are more effective

in treating the negative symptoms of schizophrenia than

antipsychotics alone.

An interesting and mostly convincing hypothesis!  Have you heard of Lurasidone?  It is quickly becoming a gold standard for treatment of bipolar I and schizophrenia both.  A literal silver bullet with a complex pharmacokinetic profile.

Lurasidone, never heard of it, but as I checked it now, it seems really good; I was researching Ziprasidone (among others: strong 5-HT1A partial agonist and 5HT2C antagonist, moderate SNRI, and relatively weak H1 antagonism) as a good candidate from the novel atypicals.