With the cost of proto PCB places and most 'modern' circuitry being SMT I haven't used one of those solderless breadboard units for well over 20 years and don't own one here in the US. I know there's one back in oz in my parents garage (where all my electronics junk ended up) from my 'youth'.
All my prototyping since being here (digital) was wirewrap and that would happily run into the 20MHz range with careful placement of the wire runs (to prevent crosstalk and other nasties).
Now everything I do is surface mount and for <$100 I can get a bunch of different protos on one panel in <1 week. Seems totally pointless to not go straight to PCB these days and not have to fight with layout issues which is important in my case since all my designs are uC based and/or typically have reasonably high current switchers.
For just daydreaming, it's easier to just throw the circuit up on LT spice to verify that it should be 'close' to working the way you think and then put it on a pcb proto run.
I sometimes have little designs that I think would be neat to try out, so I just design them up, do the PCB layout and add it to a panel. Once I have enough on a panel (maybe every month or two) I'll send the panel in to one of the PCB proto places (that doesn't charge a premium for a panelised board) and then I have a bunch of neato little things to build/test for a few weeks. I often put programming adapters for the AVR chips I use on the panel (for the 6 pins ISP header and the mating pogo pins for the PCB layout under test) and also put test adapters (again with pogo pins). This means I can have designs that don't require the 3 x 2 ISP but can have the pins spread all over the board where it makes sense from a routing perspective (most of my boards are relatively small/tight and 2 layer).
cheers,
george.