The African part of the Pan-African - Brasiliano Araçuaí-West Congo Orogen (AWCO) extends from Gabon into Angola as the NW striking West Congo Belt (WCB). It consists of a distinct eastern and western litho-structural domain. Pan-African NW folding is most evident in the eastern domain with a consistent Neoproterozoic lithostratigraphy. In the western domain poorly defined Neoproterozoic metasedimentary sequences alternate with 0.9 Ga magmatism and Palaeoproterozoic basement; they describe complexly arranged structures all over the western domain. In the Democratic Republic of Congo (DRC), the Brazzaville Congo and Angola these metasediments include the Matadi quartzites Formation (Nzadi Group). In two discrete western DRC areas, next referred to as Boma and Kimeza recent studies show that the Matadi beds are 1.12 Ga deposits intruded by 1.0 Ga granites (Baudet et al., Nseka et al., this meeting). Chronologically this fits well with respectively E3 and E4 extensional events as defined by Pedrosa-Soares and Alkmin (2011). Here we first explain how the E3 and E4 extensional events develop differently in both areas related to paleo-crustal levels, i.e. an upper (Kimeza) and a lower plate (Boma) level, separated by a low-angle detachment zone. We then claim that both events represent key positions for better understanding the extensional history, preceding the AWCO. We describe and present arguments supporting (1) that in the Boma area the 1.12 Ga Matadi beds reflect a supra-detachment sedimentary sequence on top of an amphibolite facies mylonitic foliated roof in the Palaeoproterozoic basement; (2) that crustal detachment extension culminated in extensive 1.0 Ga granite intrusions that form the Boma gneiss dome and imprint complex internal structures; (3) that, contrastingly, in the adjacent Kimeza area, the Matadi sediments unconformably overly a non-foliated greenschist facies retrograde Palaeoproterozoic basement; (4) that in its western boundary only one single 1.0 Ga granite intrusion forms a local updoming and a thermal imprint in the 1.12 Ga sediments; (5) that contrary to the complexly arranged supra-detachment Matadi beds in the Boma gneiss dome, the Matadi beds in the Kimeza area, remain nearly undeformed except nearby the single 1.0 Ga intrusion. Considering that on the scale of the AWCO, the E3 event started with listric fault-bounded 1.18 Ga Espinhaço deposits in Brazil, we conclude that the extension subsequently evolved into a detachment with the 1.12 Ga Matadi supradetachment sediments upon the lower plate and that the detachment eventually plunges under the thick, non-extended Kimeza plate boundary with unconformably overlying Matadi beds. After the E3 event the extension culminated in updoming of thinned crust by extensive 1.0 Ga E4 granite intrusions, only one of which intruded the thick Kimeza basement at its western boundary. Whereas E3-4 are confined to ductile stretching of the lower crustal segment of Boma, subsequent E5 continental rift magmatism concentrated on the lower-upper-plate boundary zone along the limb of the 1.0 Ga gneiss dome.