I already have a closed box so I think the lamp solution will work as well as a pre-heated second pan water solution.
The etching tray is transparent plastic, so I am also thinking of covering the base of the box (where the tray rests) with aluminum foil, to help keeping the solution warm.
I believe a single 60W incadescent lamp could heat the air inside the enclosed 30x30x20cm enclosure much higher than required within a few minutes. Maybe allowing for some more minutes for the solution to heat up as well.
It may be that the lamp will provide the copper onto the PCB with more heat (because metals hold heat whereas transparent liquids not so well) so that the copper is acting like a heat re-radiator, heating the liquid even more.
In all my PCBs, etching starts from the sides and progresses on the middle. On large PCBs, the lamp will be closer to the center, which will induce more heat to the center rather than the ends and this may somehow balance the etching. Experimentation is needed to verify this.
It is really very convenient the fact that I could use a standard incadescent (exposure not a problem as you mentioned), since these infrared lamps are quite expensive as I have found locally!