Hello Mike,
Yes, what you have laid out will often times present a mostly-accurate model that can be used to calculate power supply voltages, except for those cases of low voltage and high current. Notice that when we use these to calculate the output voltage of a 300V winding, we are not too concerned if the model is off by a couple of volts.
One factor that will lower your voltage and leave you a bit stumped is the voltage regulation of the power transformer. As you change from AC to DC, you begin to tax the power transformer a bit more, and your 2.5A load will be seen by the PT as something more like 3.25A, and consequently the AC output voltage will sag a bit. I don't know what this drop will be in your particular situation, but it will certainly present itself.
With the bridge rectifier, you will have two of those diodes conducting at all times, and you'll probably lose around 1V based on your load current, which is a huge percentage of the total available voltage.
With those two factors alone, the "Available" AC voltage may be more like 1.4V, and even with just the bridge and one capacitor, you may not be able to get it back up to 2.5V DC.
On the other hand, you can tuck a 5V/4A power transformer under the hood, bridge rectify that, then select an appropriate dropping resistor (or regulator) based on your results (that's whay I'd do).
-PB