good thread... it takes skill and patience to make complex schematic readable and easy to follow. i too prefer things interconnected rather than fragmented, but... in a pinch have also done the opposite when time was a factor. in particular i may use link labels on connectors/terminal blocks. makes swapping order quicker.

on the side note / off topic: i notice that your circuit uses LM2776. i have used it on several products and had trouble with it.
1. datasheet suggest charge pump cap to be 1uF but this was not good. even for quite moderate load of 20-30mA or no load at all, that value prevented IC from getting started. it looks like new caps have lower ESR that IC apparently has trouble with. so i chose 220nF later on and this problem was gone.
2. always add reverse bias low Vf Schottky diode across output. this is to prevent latch-up if supply has slow ramp. this was old remedy with similar products. if at any point output is pulled to positive voltage (don't recall threshold but it was low), internal parasitic structures cause latch-up. using low voltage Schottky clamped the output to less than 300mV and problem was gone. in normal operation this diode is reverse biased and no longer the issue,

the first issue was easily found right away and since PCBA i had to rework bunch of boards... not impressed - other TI products have nice datasheets and suggested default values always worked.
but the second one was worse as this only became known after while, when bunch of products was already shipped/installed. all but one worked... it took time to determine that the one struggler (would not start reliably every time) was since that client had "lazy" PSU (overloaded) so rise time was slow. by the time it reached rated voltage LM2776 was already latched up. not blown... just stuck. replacement units performed exactly the same at that site, but worked flawlessly in others and in the lab. TI denies that LM2776 is suffering from latch up issue but bench test results don't lie. "output on" button on bench PSU makes output turn on instantly and LM2776 start flawlessly (with 220nF cap). but gradually increasing supply voltage by hand makes it latch up - always. so add the diode or use different product.

 

Don't all the components sort of need to be interconnected together?

Of course they should. I didn’t describe the issue correctly. What I meant is something similar to schematic 2 on post 15. But with a complexity several times over.

To add to your list.
No 4-way connections. Period. If any two wires cross, they shouldn’t connected, they should only crossing over. I know a dot would indicate a connection, but over several generations of copies, the dot tends to disappear. Offset the wires a little to create a pair of T connections.

 

No 4-way connections. Period. If any two wires cross, they shouldn’t connected, they should only crossing over. I know a dot would indicate a connection, but over several generations of copies, the dot tends to disappear. Offset the wires a little to create a pair of T connections.

Agreed. I forgot that point.

 

As several others mentioned, scattering bunch of components on a paper and assigning names and letters and numbers to the legs of each components makes the schematic understandable only to the one who drew it. It would take hours to read this thing. I think looking for "Energy Star" would make more sense than an industry standard. Power supplies come in several topologies and each has pros and cons of its own, making it suitable for a specific application.