this is due to charge recombination. the thicker the layer, the more chance for charge revombination. this is the most critical factor 

I think the roots are more deep. In OTFT models are extensions of MOS models. In MOS, it is a relationship between thickness of electron/hole transport layer, we call - the channel thickness, tc, accordingly with the Ihantola-Moll model. In this model we deduce the Drain Current versus the Drain voltage and Gate voltage, in a infinitesimal portion of channel. During this demonstration, the conductivity of channel depends on the tc parameter. 

So, it is expect-able to have also the accumulation channel in Organic Cells to be dependent on the channel thickness.

I don't know if this was your question. 

this is due to charge recombination. the thicker the layer, the more chance for charge revombination. this is the most critical factor 

The increase of the thinness of the transpolar layers increases the series resistance of the solar cells but this would not affect the open circuit voltage. It may be so that the increase of the thickness while the mobility of layer is small causes the  charges to stay for longer time in the layer and hence they can recombine with oppositely charged Particles through the interface with the active material. This means that the holes in the HTL may recombine with electrons in the active layer. This reduces the the photo-current and consequently the open circuit voltage. Also increasing the thickness of the transport layer in the path of the incident solar radiation may decrease the transmittance and so the delivered photons to the active layer will be less. This may lead also to the he reduction of the photocurrent and hence the the open circuit voltage.

I believe Prof Zekry is closest here. It looks as if the thickness of the layer may be reducing the irradiance - the number of useful photons - reaching the photoactive layer which reduces the photocurrent and thus the open circuit voltage.

I also partially agree with Prof. Zekry but slightly disagree on the part concerning the series resistance effect on the Voc. if the series resistance is considerably increased in the device, the slope of the curve around Voc will also change which in extreme cases (with very high series resistances) will affect both Voc and FF.

The thickness of the film, as mentioned above, will also decrease the amount of light transmitted to the active layer which should reduce the Jsc and slightly increase the FF.

Organic solar cells, however, are not as simple as they seem to be. A lot of parameters have to be taken into consideration and one of them could also be the changes in surface morphology of the ZnO layer, especially when using the sol-gel method presented in the paper. 

In fact, this method usually leads to the formation of nanorriples and the surface roughness increases with annealing time leading to changes in defect density, charge recombination and which could also directly affect the energetic level of the ZnO layer (see http://www.nature.com/articles/srep04306).

Another aspect that, from my personal experience, often plays an important role is the surface energy of the ZnO layer prior to active layer deposition. Similarly to what is know well known as the "lotus effect", the surface energy of the ZnO surface will change with increasing surface roughness. This will change the wetting and spreading properties of the active layer solution and may lead to different vertical phase segregation in the active layer (more or less recombination).

I hope this helps!