Don.
I have tried to illustrate the first stage inlet stroke as follows( sketch below). When the piston is at the outward dead end and start to move inwards, air is drawn into the cylinder at a suction ( under)- pressure via the air inlet chamber and valves D, this air is coloured yellow, piston position 1. As the underpressure is created, the first stage outlet valves E shuts.The inwards side of the piston is now creating an overpressure ( blue) shutting valves A and opening the valves B and C, supplying air to the scavenging chamber, for C via the dotted blue hatched duct.
As the piston moves to position 2 the outward part of the piston side has still an underpressure and shuts now valve C as the piston passed the scavenging air ports, thus shutting the blue dotted hatched connection to the scavenging chamber. The inward side of the piston is having an overpressure, valves B are still open and in connection with the inward side (overpressure) of the piston, supplying air to the scavenging chamber.
When the piston is at the inwards dead end, position 3, I agree to your sketch showing a possibility of a leakage from the first stage "piston pocket" however at the dead inward end the pocket is at its max. expansion and there is hardly any pressure difference between the " pocket " and the outwards side of the piston, thus no significant leakage. Moreover a possible leakage to the 1 st. stage piston underpressure side (yellow) would be contained due to the sealing of the first stage pistonrings, the air can only enter the scavenging chamber when the outward pistonchamber pressure open valves C.
The fixing nut for the 4 stage piston rod is sealing the "pocket" towards the scavenging chamber.
Tore