Then I can't see any significant difference between your model and belief-in-belief model, which you claim to oppose.

It seems to me that what we usually call belief in belief basically means that someone holds a binary belief together with a quasi-Bayesian belief which conflicts with it.

Now it seems to be the definition of belief-in-belief, written in obscure terminology. Replace "quasi-Bayesian belief" to "actual belief", and "holds a binary belief" to "acts as if he had a belief" and that's it.