I woud use eight 3' lengths of 1" ID plastic tubing, two ahead of the wing spar and two behind the wing spar to keep the weight as centralized as possible laterally. They would carry 1.5 gallons of fuel. If less fuel is needed then the eight pieces of tubing could be shorter. That way a bank with half the fuel consumed would cause less of a lateral CG shift. With about 10 or 12 degrees of dihedral per side and a generous rudder area, control could be maintained even in a bank of a little more than 10 degrees. The drag penalty of that much dihedral is not a big consideration. It would take 7 Y-connectors and two check valves to bring the eight tank feeds down to one piece of fuel tubing. The check valves would be set up to prevent one side from feeding the other.

Throttle, elevator and rudder control would be the minimum and another channel for needle valve adjustment might be desirable.

A weight without fuel of 7 pounds seems easily attainable and 16 pounds with maximum fuel load isn't excessive. Less weight would be desirable.
With a span of 14 feet and an average chord of 12 inches, the wing loading would be 18.3 ounces per square foot. With the fuel tanks in the wing, the and a minimum fuselage crossection, the parasitic drag of the fuselage would only be one or two percent of the total drag budget. With a tail moment arm of 4" the horizontal talil area would only need to be about 12 % of the wing area and the vertical tail area about 9% of the wing area. the parasitic drag (tail, fuselage and landing gear would only be about 14 or 15 % of the low speed drag budget. A lift to drag ratio of about 12 or 15 to one would be easily attainable. The thrust requirement in level flight would only be about 1-1/3 pounds and any thing over 2-2/3 pounds of thrust would make takeoff and landing easy. Crusing speed would be about 31 feet per second fully loaded and aboiut 21 feet per second empty. Stall speed would be about 90% of crusing speed.