Yeah. The more rejection of noise that a Felicia incurs, the more heat, the more loss of power. If you have a CD player plugged into one I'd think you would want roughly double the available current at the first transformer's secondaries.
The capacitors after the second transformer are going to keep some differential mode noise out of your equipment, or out of anything else plugged into the same socket the Felicia powers. But either it gets transfered across the transformer (not too likely much will), burned up as heat, or the capacitors will indefinitely cycle it (which overtime will increase the amount of current quicker than the heat loss from your very good copper wire). I don't know which is happening... the point being current will decrease. The capacitors will continue to provide the relative amount of current they eat up, but whatever the amount of noise being rejected by them can either increase (and take more current) or cause the transformer to become less efficient as it burns off some heat. If this became too extreme the secondary transformer would actually introduce common mode noise into your equipment (which might produce its own anyways).
As long as the impedance of the transformer never leaps up and Q isn't pulled toward your equipment, the noise should stay on the other side. However inadequate sized transformers could do just that. At that point you'd be feeding all your noise in a loop, building it up faster than it burns off.
Someone is probably going to tell me I'm retarded and got it wrong but... my stuff sounds great.
